How to determine the daily streamflow of a river while fully accounting for climate change aprt from using hydrologic models such as SWAT, WEAP etc? Can the ARIMA models be twigged to suit this purpose?
how about using Lars_Wg? I generate my precipitation data with this model I think It is ok to
Yes thanks Nasrin, LARS WGN can do precipitation, but i'm not sure if it can handle future daily streamflow, any journal papers you can recommend?
rgds
Depends on the nature of problem, you can choose your desire model. In case of considering a variety of predictors (like your problem), the data-driven models (such as ANNs) can be good options, however statistical models such as time series or genetic program based methods can always be an alternative.
https://www.researchgate.net/publication/272024853_Short-Term_Forecasting_of_Water_Yield_from_Forested_Catchments_after_Bushfire_A_Case_Study_from_Southeast_Australia
Article Short-Term Forecasting of Water Yield from Forested Catchmen...
For your problem there are two types of approach. First you downscale precipitation and temperature as these two affects stream flow for your study area for the future time period and use the future generated precipitation and temperature in SWAT or other hydrologic model to find out future stream flow. Secondly you can directly forecast the stream flow for the future time period applying statistical downscaling techniques but you have to take important predictors for the future simulation. In second case no need to use any hydrologic model.
Thank you Jeew, i wish to use the second method...i will try to get more papers relating to this technique. Kind regards.
SWAT and/or Hydrological models are different from ARIMA. ARIMA needs only observed data............. but SWAT requires too much info including observed data for calibration and validation. The challenge is to obtain all info for physical-based Hydrologic models
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis